Python多进程编程执行任务

我的需求如下：现有一批任务，使用进程池执行，每个任务执行耗时不一样，任务并发执行期间，需要每隔一段时间监控任务执行进度

直接贴代码：

import multiprocessing
import time
import random
from multiprocessing import Pool, Manager


def worker(task_id, task_status, lock):
    """任务执行函数，模拟不同耗时操作"""
    try:
        # 记录任务开始状态
        with lock:
            task_status[task_id] = "running"

        # 随机生成任务执行时间（30秒到30分钟）
        execution_time = random.uniform(1, 30)
        time.sleep(execution_time)

        # 更新任务完成状态
        with lock:
            task_status[task_id] = "completed"
        return f"Task {task_id} completed"
    except Exception as e:
        with lock:
            task_status[task_id] = f"failed: {str(e)}"
        return f"Task {task_id} failed"


def monitor_tasks(task_status, total_tasks, lock):
    """任务监控函数，每30秒输出状态"""
    while True:
        time.sleep(10)
        with lock:
            running = [k for k, v in task_status.items() if v == "running"]
            completed = [k for k, v in task_status.items() if v == "completed"]
            failed = [k for k, v in task_status.items() if v == "failed"]
            remaining = total_tasks - len(running) - len(completed)

        print(f"\n=== 系统状态监控 ===")
        print(f"正在执行的任务数: {len(running)}")
        print(f"任务详情: {running}")
        print(f"剩余任务数: {remaining}")
        print(f"已完成任务数: {len(completed)}")
        print(f"失败任务数: {len(failed)}")
        print("====================")


if __name__ == '__main__':
    # 初始化管理器
    manager = Manager()
    lock = manager.Lock()

    # 创建任务状态字典（任务ID: 状态）
    task_status = manager.dict()
    total_tasks = 100

    # 初始化所有任务状态为"pending"
    for i in range(total_tasks):
        task_status[i] = "pending"

    # 创建进程池
    with Pool(processes=16) as pool:
        # 提交所有任务
        for task_id in range(total_tasks):
            pool.apply_async(worker, args=(task_id, task_status, lock))

        # 启动监控线程
        monitor_process = multiprocessing.Process(
            target=monitor_tasks,
            args=(task_status, total_tasks, lock)
        )
        monitor_process.start()

        # 等待所有任务完成
        pool.close()
        pool.join()

        # 终止监控进程
        monitor_process.terminate()
        monitor_process.join()